The present invention relates generally to electronic commerce activities and, more particularly, to a method, system, and computer program product for determining an optimal database refresh rate for a commerce system.
For e-commerce businesses, such as Web-based sales of computers, where customers intend to find and purchase products using a sales site of the Internet, along with price and function, availability lead time is one of the important pieces of information that is used by a customer to make a decision on whether to buy a product. While a customer is browsing various products in a catalog, it is important to have this information displayed immediately in order for the customer to have a positive experience. For this reason, the calculations used in determining availability should be done prior to the customer accessing the catalog and should be integrated with the display of product information.
Product availability is typically computed using a database of products and components availability and is presented during the shopping experience to set customer expectations. Based on the provided ship date, as well as other information such as price, features etc., customers make decisions on whether to purchase the product. When a customer places an order, a ship date is computed and is promised to the customer. In addition to setting an accurate availability expectation during learn, shop and buy experiences, shipping the product on time at checkout is an important indicator of customer service level. When products are shipped later than the promised ship date, customer service levels are degraded, causing an unfavorable business condition.
The availability of products and components are constantly changing as customer orders are fulfilled, production/assembly is scheduled, and components are replenished from suppliers. The size of this dynamic data is usually very large especially for large enterprises, and the data is also usually integrated with other corporate IT systems. It is also very expensive to update this database frequently as customers shop and purchase on the Internet, because of the supporting information technology (IT) architecture, such as size and number of servers and CPU time. Thus, it is a typical practice for corporations to take a snap shot of the availability outlook database a few times a day, and use this static view of the availability information to compute availability during the customer e-shopping activities. However, due to the inaccurate view of the availability in this static database, a certain number of customers will be given incorrect lead times, thus resulting in a degree of dissatisfaction.
What is needed, therefore, is a way to accurately estimate the error rate of availability lead time computation and corresponding IT expense for various refresh frequencies.